We study convex programs that involve the minimization of a convex function over a convex subset of a topological vector space, subject to a finite number of linear inequalities. We develop the notion of the quasi relative interior of a convex set, an extension of the relative interior in finite dimensions. We use this idea in a constraint qualification for a fundamental Fenchel duality result, and then deduce duality results for these problems despite the almost invariable failure of the standard Slater condition. Part II of this work studies applications to more concrete models, whose dual problems are often finite-dimensional and computationally tractable.

/pdf/partially-finite-convex-programming-part-i-quasi-relative-6q2qc3k8e5.pdf

Partially finite convex programming, Part I: Quasi relative interiors and duality theory

Using an old M. Krein’s result and a result concerning symmetric spaces from [S. Radenovic, Z. Kadelburg, Quasi-contractions on symmetric and cone symmetric spaces, Banach J. Math. Anal. 5 (1) (2011), 38–50], we show in a very short way that all fixed point results in cone metric spaces obtained recently, in which the assumption that the underlying cone is normal and solid is present, can be reduced to the corresponding results in metric spaces. On the other hand, when we deal with non-normal solid cones, this is not possible. In the recent paper [M.A. Khamsi, Remarks on cone metric spaces and fixed point theorems of contractive mappings, Fixed Point Theory Appl. 2010, 7 pages, Article ID 315398, doi:10.1115/2010/315398 ] the author claims that most of the cone fixed point results are merely copies of the classical ones and that any extension of known fixed point results to cone metric spaces is redundant; also that underlying Banach space and the associated cone subset are not necessary. In fact, Khamsi’s approach includes a small class of results and is very limited since it requires only normal cones, so that all results with non-normal cones (which are proper extensions of the corresponding results for metric spaces) cannot be dealt with by his approach.

On cone metric spaces: A survey

We prove several fixed point theorems on cone metric spaces in which the cone does not need to be normal. These theorems generalize the recent results of Huang and Zhang (2007), Abbas and Jungck (2008), and Vetro (2007). Furthermore as corollaries, we obtain recent results of Rezapour and Hamlborani (2008).

/pdf/common-fixed-point-theorems-for-weakly-compatible-pairs-on-1zvfsm36sf.pdf

Common fixed point theorems for weakly compatible pairs on cone metric spaces

The policy iteration method of dynamic programming is studied in an abstract setting. It is shown to be equivalent to the Newton-Kantorovich iteration procedure applied to the functional equation of dynamic programming. This equivalence is used to obtain the rate of convergence and error bounds for the sequence of values generated by policy iteration. These results are discussed in the context of the finite state Markovian decision problem with compact action space. An example is analyzed in detail.

On the Convergence of Policy Iteration in Stationary Dynamic Programming

In this review we describe the basic structure of positive continuous one-parameter semigroups acting on ordered Banach spaces. The review is in two parts.

Yau-Chuen Wong

Papers

Partially ordered topological vector spaces